jevbooks

← All projects

jev-ai-testing

agarwalpranav0711/jev-ai-testing

Two Jev-powered web demos: tone analyzer and maze runner.

Game & Simulation43%Runner-up: Agent Decisions

What it is

A suite of two browser demos for developers and demo viewers: Mirror, a real-time message tone analyzer scoring 8 tone dimensions, and Jev Maze Runner, a game where Jev picks paths through a random 15x15 maze.

How it uses Jev

The README does not specify which Jev primitives are used or how state is passed. It states Jev makes sub-100ms strategic path decisions in the maze and evaluates 8 tone dimensions on every keystroke pause in Mirror, with results driving the UI.

Technique worth stealing

Route browser calls through a Cloudflare Worker proxy so JEV_API_KEY stays server-side and CORS is handled.

Try it

Deploy worker.js with wrangler, set WORKER_URL, then serve mirror.html and maze.html via python -m http.server.

View on GitHub

judged by Jevjev-1.13.0

Evidence

Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.

  • Jev-centricyes0.83
  • Shows a System One patternyes0.85
  • Handles uncertaintyno0.04
  • Measuredno0.33
  • Runnableyes0.94
  • Worth recommendingunclear0.49
  • Model replicano0.03
  • Problem scopescore on a 0–2 scale0.09
  • About Jevyes0.97

Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 20 Sept 2026.